Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/zimoliao/scholaraio/auditnpx skills add ZimoLiao/scholaraio --skill auditgit clone --depth 1 https://github.com/ZimoLiao/scholaraioWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00034 | $0.00733 |
| Opus 5 | $0.00017 | $0.00367 |
| Sonnet 5 | $0.00007 | $0.00147 |
| Haiku 4.5 | $0.00003 | $0.00073 |
Grade A, and why
audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
论文审计
检查已入库论文的数据质量。分阶段：规则化检查（自动）+ LLM 深度诊断（对可疑项）+ 自动修复。
阶段一：规则化检查
scholaraio audit [--severity error|warning|info]
问题按严重程度分类：
- 错误：缺少标题、缺少 MD 文件、JSON 解析失败、DOI 重复
- 警告：缺少 DOI/摘要/年份/作者/期刊、MD 过短、标题不一致、文件名年份不匹配
- 提示：文件名不符合规范格式
阶段二：LLM 深度诊断（title_mismatch 专项）
对每篇 title_mismatch 论文，用 Read 工具读取 meta.json 和 paper.md（前 80 行），提取最像标题的候选（不只看首个 H1），判断：
- MD 正文的实际主题/标题是否与 JSON 元数据一致
- 无害（封面/前言页、MinerU 标题识别问题）vs 真正的内容错配
阶段三：修复
对确认的错配，使用 repair 命令：
# 先 dry-run 预览
scholaraio repair "<paper-id>" --title "正确标题" [--author "一作"] [--year YYYY] [--doi "10.xxx/..."] --dry-run
# 确认后执行
scholaraio repair "<paper-id>" --title "正确标题" [--author "一作"] [--year YYYY] [--doi "10.xxx/..."] [--no-api]
# 修复后重建索引
scholaraio pipeline reindex
检查规则
| 规则 | 级别 | 说明 |
|---|---|---|
missing_title |
error | 缺少标题 |
missing_md |
error | JSON 无对应 MD 文件 |
duplicate_doi |
error | DOI 重复 |
missing_doi |
warning | 缺少 DOI（document / patent 等非论文类型通常不报） |
missing_abstract |
warning | 缺少摘要 |
title_mismatch |
warning | JSON 标题与 MD 前 80 行里的标题候选不一致 |
nonstandard_filename |
info | 文件名不符合规范格式 |
补充判断口径：
paper_type为空时，仍按普通论文保守检查missing_doi/missing_journaldissertation、document、technical-report、lecture-notes等非 article 类型，优先警惕封面/前言页导致的title_mismatch误报
示例
用户说："帮我检查一下论文库有没有问题" → 执行阶段一规则化检查
用户说："深度检查" → 执行阶段一 + 阶段二（LLM 逐篇诊断 title_mismatch）
用户说："修复那些错配的论文" → 执行阶段三
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 67 lines · 34 tokens per session scan A 809d2d935a8c
audit is a skill published in the GitHub repository ZimoLiao/scholaraio (566 stars, last pushed yesterday), licensed MIT. It adds 34 tokens to every session and 733 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
research-workflow
This skill should be used when the user asks research questions, needs information lookup, wants comparisons, asks "what is", "how does", "explain", "compare", "best practices", "latest developments", or any query requiring web search, documentation lookup, or synthesis of multiple sources. Provides optimal routing…
design-lab
Scope the physical and interface design of a laboratory before configuring it. Use when a user asks how to approach a lab buildout, how to break a cell into components, which mechanism or instrument to choose, how to tailor a reference design to their material system or hazards, whether a design is over-engineered, or…
gigaxity-deep-research
Deep research MCP server wrapping Qwen3-30B-A3B-Thinking via OpenRouter. Use when an agent needs cross-source synthesis with citations, exploratory expansion of an unfamiliar topic, chain-of-thought reasoning over evidence, or fast conversational lookups grounded in live web search. Exposes six MCP tools — two…
start-here
Establish or resume an OpenSDL laboratory through a normal agent conversation. Use when a user says start here, set up or configure my lab, plan a new lab or cell, describe existing equipment, continue a lab build, or turn a laboratory idea into a simulator-first implementation plan.
orient-lab
Inspect the current OpenSDL repository and declared laboratory state. Use when a user asks for current status or needs a declared-state baseline before a concrete task; use start-here for setup planning, inventory changes, or new equipment.
release
Build and validate OpenSDL release candidates by synchronizing versions and producing every workspace distribution locally. Use when cutting or rehearsing a coordinated project release. It does not publish, sign, or tag.